An Improved Mayfly Optimization Algorithm Based on Median Position and Its Application in the Optimization of PID Parameters of Hydro-Turbine Governor

Abstract

Mayfly algorithm is a new intelligent optimization algorithm with unique optimization capabilities recently proposed. It has strong research value, but there are also insufficient explorations, and it is easy to fall into the problem of local optimization. This paper aims to improve the optimization performance of the mayfly algorithm and explore its application value in practical engineering optimization problems. An improved mayfly algorithm based on the median position of the group is proposed. In its velocity update, the median position of the group is introduced with emphasis, and a non-linear gravity coefficient is introduced at the same time. Through the benchmark test function, its superiority in exploitation, convergence speed and accuracy and the improvement of exploration are verified. At the same time, the simulation model of the hydro-turbine governor using MATLAB/Simulink is established, and 10% frequency disturbance experiments of this model are carried out separately in two typical working conditions. The experiments results show that the optimal ITAE index value of the system obtained by the improved mayfly algorithm is smaller, and 16.5 and 18.1 iterations to complete on average. In addition, the experiments results reveal that the PID parameters optimized by the improved mayfly algorithm can make the dynamic performance of the regulation system better than other popular swarm intelligence algorithms, where the overshoot decreased by more than 3.1%, and the adjustment time also decreased in different degrees. The proposal of the median position of the group provides a new idea for the improvement of the swarm intelligence optimization algorithm. Meanwhile, a new effective method for optimizing the PID parameters of the hydro-turbine governor has been found.

Integration of Fopid and Mayfly Algorithm for Sepic with Multi-Objective Functions

“…For this converter, the state space (from [36]) average modeling approach and its modifications are taken for granted. The tiny signal model equations in [36] are used to derive the differential equations for describing one switching current at the input inductor.…”

“…For this converter, the state space (from [36]) average modeling approach and its modifications are taken for granted. The tiny signal model equations in [36] are used to derive the differential equations for describing one switching current at the input inductor. It is possible to describe a discrete type of inductor in terms of the current, 𝐿1 at the beginning of the (i+1) th switching cycle; initial switching cycle (i th ) can be expressed as…”

“…Whenever both the flies are far away, they should increase their speeds at greater rates, although when they are close together, they would sustain their speeds at lower levels. Therefore, Equation (18) needs to be enhanced to accommodate for such a circumstance, as demonstrated by the following Equation (36):…”

Improved Mayfly Optimization and LightGBM Classifier for Smart City Traffic Prediction

“…al. (15) proposed an improved mayfly optimization algorithm (IMFOA) using the median position parameter and the proposed IMFOA outperformed well than the following nature inspired algorithms: Particle Swarm Optimization (PSO), Salvia Swarm Algorithm (SSA), Whale Optimization Algorithm (WOA), Genetic Algorithm (GA), and Firefly Algorithm (FA).…”

Metaheuristic Optimization Techniques Used in Controlling of an Active Magnetic Bearing System for High-Speed Machining Application

“…Guo Lei et al proposes an improved mayfly algorithm based on the median position of the group. This algorithm reduced the optimal ITAE index value of the system when used to optimize a PID parameters [21]. Izci et al developed a novel hybrid technique, arithmetic optimization algorithm-Nelder-Mead (AOA-NM) to achieve an optimum design for automobile cruise control [22].…”